As the rivalry within artificial intelligence (AI) image generation increases with the rollout of FLUX.1 and Grok-2, leading company Midjourney is rising to the challenge with a new unified AI image editor on its website.
The latest web editor combines inpainting, canvas extension, and other editing tools into a single, streamlined interface. This means that users can now repaint parts of an image with new AI-generated visuals by submitting text prompts and extending the image’s boundaries more efficiently.
Previously, these tools were scattered in more disparate menus on the website. “We think this makes editing your MJ images way more seamless than before and is a huge step forward,” Midjourney CEO David Holz said in a Discord message.
The new web editor is currently available to all users who have created at least ten images on the platform.
More updates and additions
In addition to the unified image editor, a virtual “brush”-like tool for inpainting was also debuted—an upgrade from Midjourney’s former square selector and lasso tools. This enables better precision when editing portions of an AI-generated image.
The company also launched another feature designed to enhance communication between its web and Discord communities. This feature allows messages sent in specific Web Rooms to be reflected in corresponding Discord channels and vice versa, maintaining in-sync connections across both platforms regardless of where users choose to interact.
The chat mirroring service covers rooms such as prompt-craft, general-1, and a special superuser room for users who have created more than 1,000 images.
So far, most Midjourney users have reacted positively to the changes. In a post on X, Allar Haltsonen wrote: “New Midjourney web editor is crazy,” with before and after photos of an image he edited using the editor.
New Midjourney web editor is crazy.
From left image to right one. pic.twitter.com/tizUfjqAID
— Allar Haltsonen (@AllarHaltsonen) August 16, 2024
Powering through legal battles
The latest releases come amid Midjourney’s current legal battles against a group of artists who filed a class-action lawsuit over copyright violations. The plaintiffs allege that the company used copyrighted images without consent to train its AI model.
The case is set to proceed toward the discovery phase after a judge denied Midjourney’s motion for dismissal last week. In discovery, the plaintiffs’ lawyers would be allowed access to internal company documents, which hopefully would shed light on the training practices and datasets of Midjourney and other AI firms.
Despite these hurdles, CEO Holz thanked their users for their patience during the development process. He also encouraged them to explore the additional capabilities of the editor and emphasized that more innovation and expansion will be added to improve the overall creative process and foster a more connected community among Midjourney users.